fix(video): guard ffmpeg call sites with shutil.which check

Signed-off-by: Dong Hyuk Chang <9426164+thomasdhc@users.noreply.github.com>
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: thomasdhc

nemo_curator/stages/video/clipping/clip_extraction_stages.py

@@ -14,6 +14,7 @@
 
 import copy
 import pathlib
+import shutil
 import subprocess
 import uuid
 from dataclasses import dataclass
@@ -73,6 +74,9 @@ def setup(self, worker_metadata: WorkerMetadata | None = None) -> None:  # noqa:
         Args:
             worker_metadata (WorkerMetadata, optional): Information about the worker (provided by some backends)
         """
+        if not shutil.which("ffmpeg"):
+            msg = "ClipTranscodingStage requires 'ffmpeg'. Install with: sudo apt-get install -y ffmpeg"
+            raise RuntimeError(msg)
         if self.encoder not in {"libopenh264", "libx264", "h264_nvenc"}:
             error_msg = f"Expected encoder of `libopenh264`, `libx264`, or `h264_nvenc`. Got {self.encoder}"
             raise ValueError(error_msg)

nemo_curator/stages/video/clipping/video_frame_extraction.py

@@ -12,6 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import shutil
 import subprocess
 from dataclasses import dataclass
 from pathlib import Path
@@ -38,6 +39,9 @@ def get_frames_from_ffmpeg(
     use_gpu: bool = False,
 ) -> npt.NDArray[np.uint8] | None:
     """Fetch resized frames for video."""
+    if not shutil.which("ffmpeg"):
+        msg = "get_frames_from_ffmpeg requires 'ffmpeg'. Install with: sudo apt-get install -y ffmpeg"
+        raise RuntimeError(msg)
     if use_gpu:
         command = [
             "ffmpeg",

nemo_curator/utils/windowing_utils.py

@@ -13,6 +13,7 @@
 # limitations under the License.
 
 import math
+import shutil
 import subprocess
 from dataclasses import dataclass
 
@@ -138,6 +139,9 @@ def split_video_into_windows(  # noqa: PLR0913
                 index += count
 
         if return_bytes:
+            if not shutil.which("ffmpeg"):
+                msg = "split_video_into_windows with return_bytes=True requires 'ffmpeg'. Install with: sudo apt-get install -y ffmpeg"
+                raise RuntimeError(msg)
             if len(windows) == 1:
                 return [mp4_bytes], video_frames, windows